Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758110AbaFYRdP (ORCPT ); Wed, 25 Jun 2014 13:33:15 -0400 Received: from relay.parallels.com ([195.214.232.42]:55467 "EHLO relay.parallels.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757665AbaFYRbQ (ORCPT ); Wed, 25 Jun 2014 13:31:16 -0400 Message-ID: <1403717468.3462.60.camel@tkhai> Subject: Re: [PATCH v2 1/3] sched/fair: Disable runtime_enabled on dying rq From: Kirill Tkhai To: CC: , Peter Zijlstra , Ingo Molnar , Srikar Dronamraju , Mike Galbraith , , Paul Turner , Date: Wed, 25 Jun 2014 21:31:08 +0400 In-Reply-To: References: <20140624074148.8738.57690.stgit@tkhai> <1403596432.3462.26.camel@tkhai> <53A9CB23.1080708@yandex.ru> <53A9D0F1.5080305@yandex.ru> <1403682814.3462.33.camel@tkhai> Organization: Parallels Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.8.5-2+b3 MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Originating-IP: [10.30.26.172] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org В Ср, 25/06/2014 в 09:52 -0700, bsegall@google.com пишет: > Kirill Tkhai writes: > > > В Вт, 24/06/2014 в 23:26 +0400, Kirill Tkhai пишет: > >> On 24.06.2014 23:13, bsegall@google.com wrote: > >> > Kirill Tkhai writes: > >> > > >> >> On 24.06.2014 21:03, bsegall@google.com wrote: > >> >>> Kirill Tkhai writes: > >> >>> > >> >>>> We kill rq->rd on the CPU_DOWN_PREPARE stage: > >> >>>> > >> >>>> cpuset_cpu_inactive -> cpuset_update_active_cpus -> partition_sched_domains -> > >> >>>> -> cpu_attach_domain -> rq_attach_root -> set_rq_offline > >> >>>> > >> >>>> This unthrottles all throttled cfs_rqs. > >> >>>> > >> >>>> But the cpu is still able to call schedule() till > >> >>>> > >> >>>> take_cpu_down->__cpu_disable() > >> >>>> > >> >>>> is called from stop_machine. > >> >>>> > >> >>>> This case the tasks from just unthrottled cfs_rqs are pickable > >> >>>> in a standard scheduler way, and they are picked by dying cpu. > >> >>>> The cfs_rqs becomes throttled again, and migrate_tasks() > >> >>>> in migration_call skips their tasks (one more unthrottle > >> >>>> in migrate_tasks()->CPU_DYING does not happen, because rq->rd > >> >>>> is already NULL). > >> >>>> > >> >>>> Patch sets runtime_enabled to zero. This guarantees, the runtime > >> >>>> is not accounted, and the cfs_rqs won't exceed given > >> >>>> cfs_rq->runtime_remaining = 1, and tasks will be pickable > >> >>>> in migrate_tasks(). runtime_enabled is recalculated again > >> >>>> when rq becomes online again. > >> >>>> > >> >>>> Ben Segall also noticed, we always enable runtime in > >> >>>> tg_set_cfs_bandwidth(). Actually, we should do that for online > >> >>>> cpus only. To fix that, we check if a cpu is online when > >> >>>> its rq is locked. This guarantees we do not have races with > >> >>>> set_rq_offline(), which also requires rq->lock. > >> >>>> > >> >>>> v2: Fix race with tg_set_cfs_bandwidth(). > >> >>>> Move cfs_rq->runtime_enabled=0 above unthrottle_cfs_rq(). > >> >>>> > >> >>>> Signed-off-by: Kirill Tkhai > >> >>>> CC: Konstantin Khorenko > >> >>>> CC: Ben Segall > >> >>>> CC: Paul Turner > >> >>>> CC: Srikar Dronamraju > >> >>>> CC: Mike Galbraith > >> >>>> CC: Peter Zijlstra > >> >>>> CC: Ingo Molnar > >> >>>> --- > >> >>>> kernel/sched/core.c | 15 +++++++++++---- > >> >>>> kernel/sched/fair.c | 22 ++++++++++++++++++++++ > >> >>>> 2 files changed, 33 insertions(+), 4 deletions(-) > >> >>>> > >> >>>> diff --git a/kernel/sched/core.c b/kernel/sched/core.c > >> >>>> index 7f3063c..707a3c5 100644 > >> >>>> --- a/kernel/sched/core.c > >> >>>> +++ b/kernel/sched/core.c > >> >>>> @@ -7842,11 +7842,18 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota) > >> >>>> struct rq *rq = cfs_rq->rq; > >> >>>> > >> >>>> raw_spin_lock_irq(&rq->lock); > >> >>>> - cfs_rq->runtime_enabled = runtime_enabled; > >> >>>> - cfs_rq->runtime_remaining = 0; > >> >>>> + /* > >> >>>> + * Do not enable runtime on offline runqueues. We specially > >> >>>> + * make it disabled in unthrottle_offline_cfs_rqs(). > >> >>>> + */ > >> >>>> + if (cpu_online(i)) { > >> >>>> + cfs_rq->runtime_enabled = runtime_enabled; > >> >>>> + cfs_rq->runtime_remaining = 0; > >> >>>> + > >> >>>> + if (cfs_rq->throttled) > >> >>>> + unthrottle_cfs_rq(cfs_rq); > >> >>>> + } > >> >>> > >> >>> We can just do for_each_online_cpu, yes? Also we probably need > >> >>> get_online_cpus/put_online_cpus, and/or want cpu_active_mask instead > >> >>> right? > >> >>> > >> >> > >> >> Yes, we can use for_each_online_cpu/for_each_active_cpu with > >> >> get_online_cpus() taken. But it adds one more lock dependence. > >> >> This looks worse for me. > >> > > >> > I mean, you need get_online_cpus anyway - cpu_online is just a test > >> > against the same mask that for_each_online_cpu uses, and without taking > >> > the lock you can still race with offlining and reset runtime_enabled. > >> > > >> > >> Oh, I see. Thanks. > > > > But we can check for rq->online, don't we? How about this? > > Yeah, that should work. We can't base on it because rq->offline is not available in !SMP. Could you review the message from [PATCH v3 1/3] topic? > > > > sched/fair: Disable runtime_enabled on dying rq > > > > We kill rq->rd on the CPU_DOWN_PREPARE stage: > > > > cpuset_cpu_inactive -> cpuset_update_active_cpus -> partition_sched_domains -> > > -> cpu_attach_domain -> rq_attach_root -> set_rq_offline > > > > This unthrottles all throttled cfs_rqs. > > > > But the cpu is still able to call schedule() till > > > > take_cpu_down->__cpu_disable() > > > > is called from stop_machine. > > > > This case the tasks from just unthrottled cfs_rqs are pickable > > in a standard scheduler way, and they are picked by dying cpu. > > The cfs_rqs becomes throttled again, and migrate_tasks() > > in migration_call skips their tasks (one more unthrottle > > in migrate_tasks()->CPU_DYING does not happen, because rq->rd > > is already NULL). > > > > Patch sets runtime_enabled to zero. This guarantees, the runtime > > is not accounted, and the cfs_rqs won't exceed given > > cfs_rq->runtime_remaining = 1, and tasks will be pickable > > in migrate_tasks(). runtime_enabled is recalculated again > > when rq becomes online again. > > > > Ben Segall also noticed, we always enable runtime in > > tg_set_cfs_bandwidth(). Actually, we should do that for online > > cpus only. To fix that, we check if a cpu is online when > > its rq is locked. This guarantees we do not have races with > > set_rq_offline(), which also requires rq->lock. > > > > v2: Fix race with tg_set_cfs_bandwidth(). > > Move cfs_rq->runtime_enabled=0 above unthrottle_cfs_rq(). > > v3: Check for rq->online instead of cpu_active. > > > > Signed-off-by: Kirill Tkhai > > Reviewed-by: Ben Segall -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/